Home » Fără categorie » BRAIN: Sentiment Analysis on Embedded Systems Blended Courses

Categories

Our Facebook Page

BRAIN: Sentiment Analysis on Embedded Systems Blended Courses

The paper written by Răzvan Bogdan from the Department of Computers and Information Technology, Politehnica University of Timisoara, includes a presentation of a modality of integrating Embedded Systems Massive Open Online Courses (MOOCs) into blended courses. More than that, it also provides an evaluation of this approach: the sentiment analysis technique.

Twitter sentiment analysis results
Twitter sentiment analysis results

Starting with the explanation of MOOCs, the author insists on one type of courses which is still underrepresented in the field of blended courses – that of embedded systems. Consequently, we can understand that the aim of this paper is to understand, with the help of sentiment analysis, the way in which students react to blending embedded systems MOOCs into embedded system courses. We also find out that the blending variant is applied to Embedded Systems course at “Politehnica” University of Timisoara in Romania, third year of study.

As we go on with reading the paper, we discover information about relevant previous work, for example that MOOCs evaluation has been treated in literature from differents angles: time, economical, scientific points of view; or discussing evaluating systems based on facial expression. Sentiments analysis is described as good to be used for business improving, but also the author admits that in Schouten & Frasincar is presented an algorithm which deals with aspect-level sentiment analysis, which means that the sentiment is aggregated on different entities and it is present withing the analyzed text.

However, the integration of the embedded systems Massive Open Online Courses (MOOCs) in a traditional course may be done in different modalities. On the other hand, offering a sentiment analysis research of the impact that this integration has upon students is even more important than the integration itself. The paper presents the modality which is used at “Politehnica” University of Timisoara and it consists of dividing the students in two groups: the first one dealing with the traditional approach of teaching a course and the second one dealing with a platform on which messages, assisted activities, homework, etc, are posted. The goal of integrating MOOCs in traditional Embedded Systems courses consists of broadening students’ practical perception of embedded systems intricacies. More than that, it has the aim of allowing students to become aware of the MOOC technologies.

The methodology used for doing this includes the research methods, which can be divided into two major tracks: activities pertaining to the MOOCs integration into the blended course and specific methods used to obtain the results of the sentiment analysis, each of them having specific steps that have to be followed.

In result, after applying on-site and distance-learning types of integration of MOOCs into blended courses, where 72 students were enrolled in, the author states that only 57 students, which is 79,16%, chose to complete the survey. The steps which were described in the previous section regarding the research methods were applied in order to determine the sentiment analysis from the corpus collected during the students’ survey. The results show that the polarity of the corpus is positive for all three tools: Natural Language Toolkit (NLTK), Semantria and Vivekn. In order to better understand, the author also provides three figures which illustrate the Semantria results, the Twitter sentiment analysis results and the sentiment analysis on extracted themes.

Finally, we can state that this paper presents a modality in which specific Massive Open Online Courses (MOOCs) can be integrated into a blended Embedded Systems course in a non-synchronous way. The results are very good and encouraging, because the students found the integration positive.

Read more here:

http://www.edusoft.ro/brain/index.php/brain/article/view/671

Mihaela Guţu